发表机构
Harvard University; Robert Bosch LLC Research and Technology Center（哈佛大学; 罗伯特·博世有限责任公司研究与技术中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
BranchIP通过自适应张量积计算和蒸馏损失，在保持物理精度下加速MLIPs达2.4倍并减少内存2.6倍，同时增强模型可解释性。
AI 中文摘要
等变机器学习原子间势（MLIPs）已彻底改变了原子级建模，但对复杂材料和分子系统的精确处理需要昂贵的模型。这限制了模拟的长度和时间尺度，其中张量积是关键的计算瓶颈。近期基础规模MLIPs的出现进一步加剧了这一挑战。我们提出了分支原子间势（BranchIP），一个用于学习自适应张量积计算的单模型框架，并通过新颖的蒸馏损失进行训练。在我们对两个具有物理意义的系统（一个多相催化系统和一个质子传导固体酸电解质）的实验中，BranchIP在各种模型规模上将MLIPs加速了高达2.4倍，同时将内存使用减少了高达2.6倍。这是在保持物理保真度的同时实现的。此外，学习到的自适应计算提供了模型可解释性，通过揭示哪些相互作用需要更深的计算，并展示计算深度如何与化学复杂性和动力学相关联。
英文摘要
Equivariant machine learning interatomic potentials (MLIPs) have revolutionized atomistic modeling, but accurate treatment of complex materials and molecular systems demands expensive models. This limits simulation length- and time-scales, with tensor products a key computational bottleneck. The recent emergence of foundation-scale MLIPs further exacerbates this challenge. We present Branch Interatomic Potential (BranchIP), a single-model framework for learned adaptive tensor product computation, trained with a novel distillation loss. In our experiments on two systems of physical interest, a heterogeneous catalysis system and a proton-conducting solid acid electrolyte, BranchIP accelerates MLIPs across model sizes by up to $2.4\times$ while reducing memory usage by up to $2.6\times$. This is achieved while maintaining physical fidelity. Furthermore, the learned adaptive computation provides model interpretability by revealing which interactions demand deeper computation and showing how computational depth relates to chemical complexity and dynamics.
